RESEARCH ARTICLE

Orbital complications of paranasal sinusitis

in Taiwan, 1988 through 2015: Acute
ophthalmological manifestations, diagnosis,
and management
Yi-Sheng Chang1,2*, Po-Lin Chen2,3, Jia-Horung Hung2,4, Hsiao-Yen Chen2, Chun-
Chieh Lai2,4, Chun-Yen Ou5, Chia-Ming Chang6,7,8, Chien-Kuo Wang9, Hon-Chun Cheng2,
Sung-Huei Tseng1,2*

1 Department of Ophthalmology, College of Medicine, National Cheng Kung University, Tainan, Taiwan,
2 Department of Ophthalmology, National Cheng Kung University Hospital, College of Medicine, National
Cheng Kung University, Tainan, Taiwan, 3 Department of Medicine, College of Medicine, Kaohsiung Medical
a1111111111 University, Kaohsiung, Taiwan, 4 Institute of Clinical Medicine, College of Medicine, National Cheng Kung
a1111111111 University, Tainan, Taiwan, 5 Department of Otorhinolaryngology, National Cheng Kung University Hospital,
College of Medicine, National Cheng Kung University, Tainan, Taiwan, 6 Division of Geriatrics and
a1111111111
Gerontology, Department of Internal Medicine, National Cheng Kung University Hospital, College of Medicine,
a1111111111 National Cheng Kung University, Tainan, Taiwan, 7 Division of Infectious Diseases, Department of Internal
a1111111111 Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University,
Tainan, Taiwan, 8 Center of Infection Control, National Cheng Kung University Hospital, College of Medicine,
National Cheng Kung University, Tainan, Taiwan, 9 Department of Radiology, National Cheng Kung
University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan

* willis@mail.ncku.edu.tw (YSC); shtseng1@gmail.com (SHT)

OPEN ACCESS

Citation: Chang Y-S, Chen P-L, Hung J-H, Chen H-

Y, Lai C-C, Ou C-Y, et al. (2017) Orbital
complications of paranasal sinusitis in Taiwan, Abstract
1988 through 2015: Acute ophthalmological
manifestations, diagnosis, and management. PLoS
ONE 12(10): e0184477. https://doi.org/10.1371/ Purpose
journal.pone.0184477
Paranasal sinusitis is widespread and can lead to orbital complications, blindness, and
Editor: Mohammad Shahid, Aligarh Muslim death. However, the correlation between ophthalmological findings and disease staging
University, INDIA
remains unclear. This study aimed to investigate the staging, acute ophthalmological mani-
Received: May 8, 2016 festations, diagnosis, management, and outcomes of orbital complications of paranasal
Accepted: August 24, 2017 sinusitis during a 27-year period.
Published: October 3, 2017
Methods
Copyright: © 2017 Chang et al. This is an open
access article distributed under the terms of the We retrospectively reviewed the medical records of all patients with orbital complications
Creative Commons Attribution License, which of paranasal sinusitis hospitalized at the National Cheng Kung University Hospital, a med-
permits unrestricted use, distribution, and
ical center in Taiwan during 1988–2015. Sex, age, symptoms, history, ophthalmological
reproduction in any medium, provided the original
author and source are credited. findings, laboratory and imaging findings, treatments, and outcomes were analyzed by
staging.
Data Availability Statement: All relevant data are
within the paper.
Results
Funding: The author received no specific funding
for this work. Eighty-three patients aged 9 days to 80 years had stage I (preseptal cellulitis, n = 39
Competing interests: The authors have declared patients), II (postseptal orbital cellulitis, n = 8), III (subperiosteal abscess, n = 16), IV (orbital
that no competing interests exist. abscess, n = 8), or V (intracranial involvement, n = 12) complications. Peak incidences

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 1 / 14

 Orbital complications of sinusitis

occurred in patients aged 0–19 and 60–69 years. Chronic sinusitis and diabetes mellitus
were common preexisting diseases. Extraocular movement limitation and proptosis pre-
dicted postseptal (stage II or more) involvement. The likelihood of elevated intraocular pres-
sure increased with stage. Reduced visual acuity and presence of relative afferent pupillary
defect indicated consideration of magnetic resonance imaging to investigate possible intra-
cranial extension. Ipsilateral maxillary (81.7%) and ethmoidal (75.6%) sinuses were the
most common sources of infection, and the most frequently implicated pathogens were
coagulase-negative Staphylococcus spp. (25.3%) and Staphylococcus aureus (20.5%). All
patients received intravenous antimicrobial therapy (multi-drug therapy in 88.0%), and
55.4% underwent surgery, most commonly endoscopic sinus surgery. One (1.2%) diabetic
man with stage V complications died of fungal sinusitis with intracranial invasion. Five
(6.0%) patients, all stage V, lost vision despite intensive treatment. The average length of
hospital stay was 13.8 days (range 2–72 days), and significantly longer stays were associ-
ated with stages II–V as compared to stage I.

Conclusions
Orbital infection originating from paranasal sinusitis can cause vision loss and death due to
intracranial extension. Acute ophthalmological findings predict staging and prognosis.
Cooperative consultation between ophthalmologists, otorhinolaryngologists, and neurolo-
gists is essential. Urgent diagnostic studies and aggressive antimicrobial therapy are indi-
cated, and surgery should be considered.

Introduction
Paranasal sinusitis is widespread, with an annual prevalence of 13.0–16.0% in the United States
[1], 10.9% (range 6.9–27.1%) in Europe [2], and 8.0% (range 4.2–10.2%) in China [3]. Its most
common complication, orbital infection, occurs when pathogens pass from an infected maxil-
lary, ethmoidal, frontal, or sphenoidal sinus into the orbit, either directly though neurovascu-
lar foramina or a congenital or acquired bony dehiscence, or indirectly through valveless veins
of the sinuses and orbit.
The process involves edema of the sinus mucosa, which narrows the ostia and impairs sinus
drainage. Bacterial or fungal microflorae in the sinuses proliferate and invade the edematous
mucosa, resulting in suppuration. It is augmented by reduced oxygen tension within an
obstructed sinus cavity. Then these organisms enter the orbit, leading to preseptal or orbital
inflammation. Moreover, subperiosteal or orbital abscesses may occur. The resulting elevation
of intraorbital pressure results in periorbital swelling, proptosis, ophthalmoplegia, chemosis,
and optic nerve compression. Furthermore, some may extend to the brain leading to inflam-
mation, abscess formation, or cavernous sinus thrombosis.
Although bacteria are more commonly seen in acute sinusitis. fungal infections usually
occur in immunocompromised individuals. While antimicrobial therapy has reduced
the risks of permanent sequelae, orbital involvement can still lead to blindness, or where
intracranial extension occurs, death. We aimed to investigate the clinical features of
orbital complications of sinusitis in a Taiwanese population, and observed that certain
ophthalmological manifestations and outcomes were significantly associated with disease
stage.

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 2 / 14

 Orbital complications of sinusitis

Patients and methods

Patients
Medical records of all patients diagnosed with orbital complications of sinusitis hospitalized at
the National Cheng Kung University Hospital (NCKUH) in Taiwan between January 1988
and December 2015 were reviewed. The inclusion criteria of our study were inpatients of all
age groups presenting with preseptal/orbital inflammation secondary to acute (<1 month),
subacute (1–3 months) or chronic sinusitis (>3 month) of bacterial, fungal or unidentified
pathogens. The diagnosis was based on clinical symptoms and signs as well as laboratory tests
and radiological investigations. Sinusitis was confirmed by the presence of sinus opacification
or air-fluid levels on computerized tomography (CT) or X-ray when CT was not available. The
exclusion criteria were hematological bone marrow disorders and sinus/orbital cancers. The
data recorded included sex, age, symptoms, medical history, ophthalmic examination findings,
systemic and laboratory findings, imaging results and the sinuses involved, culture results,
treatments, hospital stay durations, and outcomes. This retrospective study was approved by
the Institutional Review Board of NCKUH (A-ER-102-336), which waived the need for
informed consent because patient anonymity was inherent in the data source.

Staging
In this study, orbital complications of sinusitis were classified via a modified form of the classi-
fication system reported by Chandler et al [4]. Focal thickening and infiltration of the eyelid
anterior to the orbital septum were classified as stage I, preseptal cellulitis. Edema and inflam-
mation of the orbital contents without evidence of abscess formation were classified as post-
septal stage II, orbital cellulitis. Abscess formation between the orbital wall and the periorbita
was classified as stage III, subperiosteal abscess. Abscess formation with pus or debris within
the orbital content was classified as stage IV, orbital abscess. Additional intracranial extension,
including cavernous sinus thrombosis, meningitis, cerebritis, or epidural/subdural/intracere-
bral abscess or empyema, was classified as stage V [5–7].

Statistics
Categorical variables were analyzed using the Chi-square test or Fisher’s exact test, and contin-
uous variables were analyzed using Student’s t-test, via SPSS software (version 20, IBM,
Armonk, New York, USA). A P value < 0.05 was considered statistically significant.

Results
Eighty-three patients (35 male, 48 female) aged 9 days to 80 years (mean 33.7, SD 26.4 years)
were identified. Younger patients were most commonly afflicted, with 24 (28.9% of the total)
aged 0–9 years, followed by 12 (14.5%) patients aged 10–19 years, though notably there was an
additional peak at 60–69 years (10, 12.0%) (Fig 1).

Clinical findings
Table 1 presents clinical findings classified by stage of sinusitis-related orbital complications.
Thirty-nine (47.0%) of the 83 patients had preseptal disease (stage I), 32 (38.6%) had postseptal
disease (stages II–IV), and 12 (14.5%) had intracranial involvement (stage V). Seventy-one
(85.5%) presented symptoms typical of upper respiratory tract infections (e.g., nasal obstruc-
tion, mucopurulent drainage, headache, fever, fatigue, or cough) before admission, and 5
(6.2%) had experienced traumatic blowout fracture in the last 3 weeks. Thirty (36.1%) had a
history of chronic sinusitis (45.8% in adults versus 22.9% in children; P = 0.03), 20 (24.1%) had

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 3 / 14

 Orbital complications of sinusitis

Fig 1. Age distribution of 83 patients with orbital complications of sinusitis. Age peaked at 0–9 years, followed by 10–19 years. An
additional peak occurred at 60–69 years.
https://doi.org/10.1371/journal.pone.0184477.g001

diabetes mellitus, 3 (3.6%) adults had chronic illnesses (including end stage renal disease, hep-
atitis C-related liver cirrhosis, or poliomyelitis-related paraplegia), 2 (2.4%) children had
asthma, and 1 (1.2%) child had an atrial septal defect. Diabetes mellitus was associated with
disease stage, with 10.3% in stage I, 28.1% in stages II–IV, and 58.3% in stage V (P = 0.006,
stages II–V versus stage I), but chronic sinusitis was borderline with regard to significance
(P = 0.06).
All patients had periorbital swelling. As shown in Table 1 and Fig 2, clinical findings that
predicted more severe disease (stages II–V versus stage I) were limitation of extraocular move-
ment (EOM), proptosis, elevated intraocular pressure (IOP), reduced visual acuity, and relative
afferent pupillary defect (RAPD) (all Ps < 0.01). The changes in these ophthalmological
parameters with stage were observed in three patterns. First, the incidences of EOM limitation
and proptosis substantially increased from stage II, and thereafter were sustained at high per-
centages in stages III–V. Second, the incidence of IOP elevation increased steadily with stage.
Third, the incidences of reduced visual acuity and RAPD substantially increased at stages IV
and V respectively.
Fever was present in 24 (68.6%) of 35 children ( 19 years) versus 21 (43.8%) of 48 adults
(P < 0.001), and was observed in only 7 (35.0%) of those aged 60 years or older (P = 0.006; chil-
dren versus those aged 60 years or older). Leukocytosis and elevated C-reactive protein were
seen in 68.7% and 81.9% of subjects respectively, but were not associated with disease severity.
CT of orbits and sinuses was conducted in 71 (85.5%) patients, and sinusitis and preseptal/
orbital inflammation was evident in all these cases. The remaining 12 patients (10 children and
2 adults, all stage I), underwent only Waters’ view X-ray. Magnetic resonance imaging (MRI)
was performed in 11 (13.3%) patients, all stage V, due to suspected intracranial extension. As
shown in Table 1, the ipsilateral maxillary (68 patients, 81.9%) and ipsilateral ethmoidal (63
patients, 75.9%) sinuses were the most commonly involved. Twenty (24.4%) patients had only
1 sinus involved, and the rest had 2–6 sinuses involved. The involvement of more sinuses was

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 4 / 14

 Orbital complications of sinusitis

Table 1. Demographic characteristics and clinical findings in 83 patients by stage of orbital complications of sinusitis.
Stage, Number (%)
Preseptal Postseptal Intracranial
I II III IV V Total
Characteristics / finding 39 (47.0) 8 (9.6) 16 (19.3) 8 (9.6) 12 (14.5)b 83 (100) P valuea
Sex 0.81
Male 17 (43.6) 3 (37.5) 6 (37.5) 2 (25.0) 7 (58.3) 35 (42.2)
Female 22 (56.4) 5 (62.5) 10 (62.5) 6 (75.0) 5 (41.7) 48 (57.8)
Age group 0.49
Adult 21 (53.8) 7 (87.5) 8 (50.0) 4 (50.0) 8 (66.7) 48 (57.8)
Child ( 19 years) 18 (46.2) 1 (12.5) 8 (50.0) 4 (50.0) 4 (33.3) 35 (42.2)
Age (years)
Mean ± SD 30.0 ± 26.6 54.7 ± 19.9 31.3 ± 25.1 25.6 ± 28.5 40.3 ± 28.0 33.7 ± 26.4 0.55
Range 1–80 0–77 2–73 2–69 0–73 0–80
URI-like symptoms 32 (82.1) 8 (100.0) 15 (93.8) 6 (75.0) 10 (83.3) 71 (85.5) 0.39
History
Chronic sinusitis 10 (25.6) 4 (50.0) 7 (43.8) 4 (50.0) 5 (41.7) 30 (36.1) 0.06
Diabetes mellitus 4 (10.3) 5 (62.5) 3 (18.8) 1 (12.5) 7 (58.3) 20 (24.1) 0.006
Ophthalmological finding
EOM limitation 6 (15.4) 6 (75.0) 13 (81.3) 6 (75.0) 9 (75.0) 40 (48.2) < 0.001
Proptosis 4 (10.3) 5 (62.5) 10 (62.5) 6 (75.0) 7 (58.3) 32 (38.6) < 0.001
IOP > 23 mmHg 4 (10.3) 2 (25.0) 6 (37.5) 3 (37.5) 5 (41.7) 20 (24.1) 0.006
VA changec 3 (7.7) 1 (12.5) 3 (18.8) 4 (50.0) 10 (83.3) 21 (25.3) 0.001
RAPD present 0 0 1 (6.3) 1 (12.5) 6 (50.0) 8 (9.6) 0.005
Systemic finding
Fever (> 37.5˚C) 19 (48.7) 3 (37.5) 9 (56.3) 6 (75.0) 8 (66.6) 45 (54.2) 0.34
Leukocytosis (> 10,000/mm3) 27 (69.2) 4 (50.0) 10 (62.5) 6 (75.0) 10 (83.3) 57 (68.7) 0.91
CRP > 10.0 mg/L 31 (79.5) 8 (100.0) 12 (75.0) 7 (87.5) 10 (83.3) 68 (81.9) 0.59
Sinusitisd
Maxillary
Ipsilateral 31 (79.5) 5 (62.5) 12 (75.0) 8 (100.0) 12 (100.0) 68 (81.9) 0.59
Contralateral 14 (35.9) 1 (12.5) 3 (18.8) 3 (37.5) 5 (41.7) 26 (31.3) 0.40
Ethmoidal
Ipsilateral 24 (61.5) 7 (87.5) 14 (87.5) 8 (100.0) 10 (83.3) 63 (75.9) 0.004
Contralateral 8 (20.5) 1 (12.5) 3 (18.8) 3 (37.5) 7 (58.3) 22 (26.5) 0.24
Frontal
Ipsilateral 15 (38.5) 0 11 (68.8) 3 (37.5) 5 (41.7) 34 (41.0) 0.66
Contralateral 6 (15.4) 0 3 (18.8) 1 (12.5) 1 (8.3) 11 (13.3) 0.59
Sphenoidal
Ipsilateral 4 (10.3) 0 0 3 (37.5) 4 (33.3) 11 (13.3) 0.45
Contralateral 2 (5.1) 0 0 2 (25.0) 3 (25.0) 7 (8.4) 0.31

Abbreviations: CRP, C-reactive protein; EOM, extraocular movement; IOP, intraocular pressure; RAPD, relative afferent pupillary defect; SD, standard
deviation; URI, upper respiratory tract infection; VA, visual acuity.
a
Comparison between stage I and advanced stages (II–V).
b
Stage V: cavernous sinus thrombosis in 5 patients, frontal lobe abscesses in 4, and meningitis in 3.
c
VA change: reduction in visual acuity of  2 lines compared with fellow eye or baseline.
d
Imaging evidence of sinusitis, by side of ocular complications

https://doi.org/10.1371/journal.pone.0184477.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 5 / 14

 Orbital complications of sinusitis

Fig 2. Percentages of acute ophthalmological findings in 83 patients by stage of orbital complications of sinusitis. The increasing
incidences of acute ophthalmological findings by stage were categorized into three patterns. “EOM limitation” and “proptosis” substantially
increased from stage II, and thereafter were sustained at high percentages during stages III–V. “IOP > 23 mmHg” increased steadily by stage.
The incidences of “VA change” and “RAPD present” substantially increased at stages IV and V respectively. Abbreviations: EOM,
extraocular movement; IOP, intraocular pressure; RAPD, relative afferent pupillary defect; VA, visual acuity.
https://doi.org/10.1371/journal.pone.0184477.g002

associated with more severe orbital disease stages (Fig 3). Fig 4 shows the characteristic imag-
ing findings for each stage of orbital complications.
Cultures from the sinuses, orbit, or brain in 48 (57.8%) patients yielded the growth of a sin-
gle pathogen (25 cases; 30.4%) or polymicrobial pathogens (23 cases; 27.7%) (Table 2). The
most frequently implicated pathogens were coagulase-negative Staphylococcus spp. (25.3%)
and Staphylococcus aureus (20.5%). Blood cultures were positive in 6 (7.3%) patients, including
three aged 9 days, 23 days, and 2 years.

Treatments
All patients received intravenous antimicrobials, 9 (10.8%) with amoxicillin/clavulanate alone,
1 (1.2%) with ampicillin/sulbactam alone, and the other 73 (88.0%) with multiple antimicrobi-
als. The most frequently administered antimicrobials were amoxicillin/clavulanate used in 45
(54.2%) patients, gentamicin in 25 (30.1%), and ampicillin/sulbactam in 21 (25.3%).
Surgery was performed in 46 (55.4%) patients (Table 3), including endoscopic sinus sur-
gery in 38 (45.8%), orbitotomy in 3 (3.6%), frontoethmoidectomy in 2 (2.4%), craniotomy in 2
(2.4%), and Caldwell-Luc antrostomy in 1 (1.2%). Surgery was performed in 16 (34.0%) of the
patients with stage I–II complications versus 30 (83.3%) of those with stage III–V complica-
tions (P < 0.001), and in 19 (54.3%) children versus 27 (56.3%) adults (P = 0.86). The mean
(± SD) time from admission to surgery was 2.4 (± 5.6) days (range 0–21 days).

Outcomes
The mean (± SD) hospital stay was 13.8 (± 13.9) days (range 2–72 days) (Table 3), and was
significantly longer in cases of more severe disease (P < 0.001; stages II–IV versus stage I).

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 6 / 14

 Orbital complications of sinusitis

Fig 3. Percentages of sinuses involved in 83 patients by stage of orbital complications of sinusitis. There was a trend of a positive
association between stage and the number of sinuses involved. The most frequently involved numbers of sinuses were 1 for stages I and
II, 3 for stage III, and 4 or more for stages IV and V.
https://doi.org/10.1371/journal.pone.0184477.g003

Fig 4. Characteristic findings of each stage of sinusitis-related orbital complications determined via computed tomography (CT) or magnetic
resonance imaging (MRI). (A) Stage I, preseptal cellulitis (arrow) on an axial CT image. (B) Stage II, orbital cellulitis (arrow) on an axial CT image. (C)
Stage III, subperiosteal abscess (arrow) on a sagittal CT image. (D) Stage IV, orbital abscess (arrow) with a tiny air bubble on an axial CT image. (E and F)
Stage V, cavernous sinus thrombosis (arrow) on a coronal contrast-enhanced T1-weighted MRI, and focal cerebritis (arrowheads) on a coronal contrast-
enhanced T1-weighted MRI and an axial T2-weighted MRI. * Ethmoidal sinusitis. † Artifact due to metal material. Dashed lines: Proptosis in one eye
compared with the fellow eye.
https://doi.org/10.1371/journal.pone.0184477.g004

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 7 / 14

 Orbital complications of sinusitis

Table 2. Pathogens isolated from the sinuses, orbit, or brain in 48 of 83 patients with orbital complica-
tions of sinusitis.
Pathogen Number (% of 83) patientsa
Aerobes
Coagulase-negative Staphylococcus spp. 21 (25.3)
Staphylococcus aureus 17 (20.5)
Pseudomonas aeruginosa 6 (7.2)
Viridans Streptococcus group 6 (7.2)
Streptococcus pneumoniae 5 (6.0)
Klebsiella spp. 4 (4.8)
Enterococcus spp. 4 (4.8)
Bacillus spp. 4 (4.8)
Enterobacter spp. 3 (3.6)
Neisseria spp. 2 (2.4)
Escherichia coli 2 (2.4)
Acinetobacter baumanii 1 (1.2)
Citrobacter diversus 1 (1.2)
Flavobacterium spp. 1 (1.2)
Morganella moganii 1 (1.2)
Anaerobes
Prevotella spp. 3 (3.6)
Peptostreptococcus micros 2 (2.4)
Propionibacterium spp. 2 (2.4)
Haemophilus influenzae 2 (2.4)
Fusobacterium spp. 1 (1.2)
Gemella spp. 1 (1.2)
Fungi
Yeast 5 (6.0)
Aspergillus spp. 3 (3.6)
a
A single pathogen in 25 cases, and polymicrobial infection in 23 cases.

https://doi.org/10.1371/journal.pone.0184477.t002

Recovery was graded as complete in 76 (92.7%) patients, including 3 children too young to
measure visual acuity whose periorbital inflammation resolved completely. A 60-year-old dia-
betic man with stage V disease presented with fungal sinusitis and orbital complications, and
died of intracranial invasion to the left cavernous sinus and left medial temporal lobe within
15 days of hospitalization. One stage III patient with preexisting paraplegia due to poliomyeli-
tis died of recurrent episodes of pneumonia and respiratory failure considered unrelated to
ophthalmic complications of sinusitis. Five patients (all stage V) lost vision permanently in the
affected eye, despite intensive treatment.

Discussion
To our knowledge, this is the first study analyzing disease stage and risk factors, manifesta-
tions, diagnosis, treatments, and outcomes associated with orbital complications of sinusitis
conducted in an Asian population. We detected incidence peaks in patients aged 0–19 and
60–69 years. Chronic sinusitis and diabetes mellitus were common preexisting diseases. EOM
limitation and proptosis predicted postseptal involvement, and reduced visual acuity and the
presence of RAPD indicated that intracranial extension was likely. Maxillary and ethmoidal
sinuses were common sources of infection, and the number of sinuses involved tended to

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 8 / 14

 Orbital complications of sinusitis

Table 3. Treatments and outcomes in 83 patients by stage of orbital complications of sinusitis.

Stage, Number (%)
Preseptal Postseptal Intracranial
I II III IV V Total
Treatment / outcome 39 (47.0) 8 (9.6) 16 (19.3) 8 (9.6) 12 (14.5) 83 (100) P valuea
Treatment
Number (%) undergoing surgery 13 (33.3) 3 (37.5) 13 (81.3) 6 (75.0) 11 (91.7) 46 (55.4) < 0.001
Days from admission to surgery
Mean ± SD 1.3 ± 1.7 2.0 ± 1.2 1.9 ± 1.7 2.0 ± 1.6 7.1 ± 7.0 2.4 ± 5.6 0.11
Range 0–6 0–4 0–2 0–5 1–21 0–21
Days of hospitalization
Mean ± SD 7.0 ± 3.8 22.1 ± 18.6 11.2 ± 4.3 19.1 ± 18.9 30.2 ± 22.5 13.8 ± 13.9 < 0.001
Range 2–20 5–49 5–22 5–62 6–72 2–72
Outcome
Death 0 0 1 (6.3)b 0 1 (8.3)c 2 (2.4) 0.18
Blindness 0 0 0 0 5 (41.7)d 5 (6.0) 0.03

Abbreviations: CRP, C-reactive protein; EOM, extraocular movement; IOP, intraocular pressure; RAPD, relative afferent pupillary defect; SD, standard
deviation; URI, upper respiratory tract infection; VA, visual acuity.
a
Comparison between stage I and advanced stages (II–V).
b
Cause of death was recurrent episodes of pneumonia and respiratory failure during admission, due to preexisting chronic immobility resulting from long-
standing paralysis, and was considered unrelated to orbital complications of sinusitis.
c
Cause of death was intracranial invasion to the left cavernous sinus and left medial temporal lobe, resulting from invasive fungal sinusitis and orbital
complications.
d
No light perception in 2 patients and counting fingers in 3 patients.

https://doi.org/10.1371/journal.pone.0184477.t003

increase with stage. Staphylococci were common pathogens. More than a half of patients
underwent surgery. Among the 12 patients with stage V complications, 1 died and 5 lost vision
in the affected eyes.

Age at diagnosis and preexisting diseases

Acute sinusitis is the most common cause of orbital infections in children [8], and we found a
peak incidence of orbital complications in this age group. This is concordant with results
reported by Ferguson and McNab [9], and may be attributable to incomplete paranasal sinus
development and thinner bony barriers in such patients. In contrast, acute exacerbation of
chronic sinusitis is often the cause of serious complications in adults, and a history of chronic
sinusitis was identified in 45.8% of adults versus 22.9% of children in this study. Predisposing
factors of sinusitis include anatomic derangements, impaired sinus drainage, and inhibition of
mucociliary transport, which promote bacterial overgrowth [10]. In addition, immunocom-
promised states (e.g., diabetes mellitus, chronic renal failure, chronic liver disease, high-dose
steroid therapy, or acquired immune deficiency syndrome) reportedly predispose patients to
bacterial sinusitis and orbital cellulitis [5]. In the current study, diabetes mellitus was present
in 24.1% of patients, liver cirrhosis in 1.2%, and end stage renal disease in 1.2%. We attributed
the observed peak in the patients aged 60–69 (and older) to immunosuppression due to aging,
diabetes mellitus, or other chronic systemic diseases.

Ophthalmological but not systemic findings as indicators of disease

severity
In addition to the periorbital swelling observed in all patients in this study, acute ophthalmo-
logical findings complicated by sinusitis included EOM limitation, proptosis, IOP elevation,

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 9 / 14

 Orbital complications of sinusitis

reduced visual acuity, and RAPD. More severe orbital sinusitis complications were also signifi-
cantly correlated with these ophthalmological conditions, which is concordant with previously
reported studies [9,11–13]. A notable contribution of this study is that we categorized increases
in these ophthalmological manifestations by stage into three patterns in an effort to investigate
their predictive value with regard to disease staging and prognosis. Generally, the manifesta-
tion of only periorbital swelling predicts stage I, while concurrent EOM limitation or proptosis
predicts stage II [8]. Additionally, elevation of IOP renders stage III more likely. Visual
impairment predicts stage IV, and the development of RAPD raises a higher possibility of
stage V, which warrants further intracranial imaging and neurological assessment, and pre-
dicts potential blindness and death.
With regard to systemic manifestations, leukocytosis and elevation of C-reactive protein
were common in the current study (68.7% and 81.9% respectively), indicating an infected or
inflamed state. However, only 54.2% of patients had fever on presentation, and we surmised
that some had likely been treated with antimicrobials and/or antipyretic agents by local physi-
cians prior to presentation, as symptoms typical of upper respiratory tract infections were
present in most. Notably, the absence of fever does not exclude severe infection, especially in
those with reduced immune function due to aging or chronic disease [14]. Our finding of
fever in 65.7% of children compared to 43.8% of adults and 35.0% of patients aged 60 years or
older is concordant with the results of a study reported by Ferguson and McNab [9], in which
71% of children versus 32% of adults had temperatures > 37.5˚C.

Diagnostic imaging
CT scanning was performed in 85.5% of our patients, and accurately established the diagnosis
of sinusitis and the extent of orbital complications. Some have suggested that CT scanning
should be reserved for cases in which a postseptal orbital disease is suspected, after 24 hours of
medical treatment without improvement, or when vision cannot be accurately assessed [7].
However, Rubin et al. [15] have reported that two patients with sinusitis-related subperiosteal
abscess only presented symptoms of eyelid edema and erythema. This supports our recom-
mendation that CT scanning be performed to exclude or confirm sinusitis, either with or with-
out orbital involvement, especially in patients with a history of sinusitis and no history of
eyelid trauma. CT is also useful in the planning of surgery, because it delineates extraocular
muscles, the optic nerve, orbital walls, and the bony margins of nearby sinuses [6,8,16]. MRI
on the other hand, which we performed in 11 of 12 patients with intracranial extension, better
delineates intracranial complications such as cavernous sinus thrombosis, meningitis, cerebri-
tis, and epidural/subdural/intracerebral abscess or empyema [6,16].
Our finding that the ipsilateral maxillary and ethmoidal sinuses were the most common
sources of infection causing orbital/intracranial complications is concordant with previous
studies [11,13,16]. Infection in the ethmoidal sinus may spread directly into the orbit via the
thin bone of the lamina papyracea, or indirectly to the brain via septic thrombophlebitis [6].
Ipsilateral frontal sinusitis occurred in 34 (41.0%) of the patients in the current study, but was
not observed in those younger than 8 years; which is logical as the frontal sinus does not
develop until after the age of 6 years. Notably however, infection in the frontal sinus spreading
to the brain via the thin bone of the anterior cranial fossa, resulting in frontal lobe abscess, is
reportedly the most common intracranial complication of sinusitis [16].

Pathogens
Pathogens isolated from our patients’ sinuses, orbit, or brain were most often coagulase-nega-
tive Staphylococcus spp. and Staphylococcus aureus, which is concordant with previous reports

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 10 / 14

 Orbital complications of sinusitis

[5,9,13]. Anaerobes were isolated in only 5 (6.0%) of our cases, and were usually associated
with polymicrobial infection, as has been described in some previous reports [9,17,18]. Harris
[19] reported that no growth or growth of a single aerobic species was likely in cultures from
patients aged younger than 9 years, and that polymicrobial or anaerobe growth was more likely
with increasing age after 9 years. A notable medical improvement that has occurred over time
is that Haemophilus influenzae type B, Streptococcus pneumoniae, and Viridans Streptococcus
group previously often caused sinusitis or orbital infection in young children, but they are no
longer significant causes of such infections following the introduction of relevant vaccines in
the years 1985 and 2000 respectively [7,20]; as confirmed by the results of the current study.
Sinus or orbital infection can lead to bacteremia, which is reportedly more common in
newborns and infants than in older subjects due to their immature immune systems [21]. This
contention is consistent with our finding that 3 of our 6 patients with positive blood cultures
were either newborns or young children. In fact, orbital complications of acute sinusitis are
very rare in infants, and to date only 11 cases aged 10–74 days have been reported in the litera-
ture in the past 50 years [21]. The current study adds 2 cases to this historically documented
cohort: one was aged 23 days and exhibited stage II complications, and the other was aged 9
days and exhibited stage V complications. Both were successfully treated and recovered.

Treatment
While recommendations of antimicrobial treatment for orbital complications of sinusitis vary,
most clinicians suggest multi-drug combinations or a single broad-spectrum antimicrobial, to
safe-guard against polymicrobial pathogens including anaerobes [7,9]. Some agree that amoxi-
cillin/clavulanate, which has historically often been used (alone or in combination) and was
used in our study is effective against beta-lactamase-producing aerobes and anaerobes, is suit-
able for use in all age-groups [19].
The goals of surgery for orbital complications of sinusitis are to drain the abscess ade-
quately, release pressure in the orbit, and obtain material for culture.7 Endoscopic sinus sur-
gery, introduced in the 1980s [22], was the most frequently used procedure in the current
study. It has several advantages over an open procedure, including the negation of an external
wound, less postoperative edema, and more rapid recovery [18]. The likelihood of surgery
increased with more advanced stages of orbital complications in the current study.
It is generally believed that preseptal cellulitis and orbital cellulitis respond to drug treat-
ment alone, but subperiosteal/orbital abscess or intracranial complications require surgical
drainage [7–9,15]. Several reports indicate that drug treatment alone may be effective for eth-
moidal sinusitis-related subperiosteal abscess in some children aged younger than 9 years with
intact visual function, provided they meet certain additional criteria [19,23,24]. Nevertheless,
good clinical judgment should always take precedence and emergency drainage of a subperios-
teal abscess may be necessary.

Outcomes and prognosis

Orbital complications of sinusitis can lead to blindness and death. Causes of vision-loss
include: (1) optic neuritis resulting from a reaction to an adjacent or nearby infection, (2)
ischemia resulting from thrombophlebitis along the valveless orbital veins, and (3) pressure
ischemia, possibly causing central retinal artery occlusion [15–17]. Prompt decompression of
the orbit in such cases may preserve the patient’s vision. The rate of blindness in the current
study was 6.0%, which is statistically comparable with that of 2.5% reported by Patt and Man-
ning [25]. Notably, blindness was exclusively and commonly (41.7%) observed in conjunction
with stage V disease in the current study. Importantly, death resulting from orbital

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 11 / 14

 Orbital complications of sinusitis

complications of sinusitis is rare if the infection is appropriately treated [21,26]. However,

stage V disease can deteriorate rapidly and lead to death, as occurred in the patient in the cur-
rent study who expired within 15 days of hospitalization.

Study advantages and limitations

This study has the advantages of being a large case series of orbital complications of sinusitis
encompassing five stages, in which the risk factors, manifestations, diagnosis, treatment, and
outcomes were analyzed statistically in conjunction with stage. Most importantly, this study
suggests useful ophthalmological indicators for assessing disease staging and prognosis. One
limitation of the current study is that we only included hospitalized patients. The exclusion
of those with mild preseptal cellulitis that were treated on an outpatient basis unavoidably
resulted in an underestimation of the relative frequency of this stage. Another limitation is the
retrospective design, which is inherently subject to inaccuracies or inconsistencies in terms of
the data obtained and the manner in which they are interpreted or documented.

Conclusion
Orbital complications of sinusitis tended to occur in children and aged individuals, and diabe-
tes mellitus predisposed patients to more severe disease stages. EOM limitation and proptosis
predicted postseptal involvement, and reduced visual acuity and the presence of RAPD indi-
cated the likelihood of intracranial extension which poses a risk of blindness and death. Coop-
erative consultation between ophthalmologists, otorhinolaryngologists, and neurologists is
advisable. Urgent CT or MRI to delineate the extent of disease, intravenous antimicrobial ther-
apy adjusted based on culture results, and prompt surgical drainage when indicated are essen-
tial to an optimal prognosis.

Acknowledgments
The authors thank Zhao-Hong Cheng, Ph.D. (Department of Business Management, National
Sun Yat-sen University, Kaohsiung, Taiwan) for assistance with the statistical analysis.

Author Contributions
Conceptualization: Yi-Sheng Chang, Sung-Huei Tseng.
Data curation: Yi-Sheng Chang, Po-Lin Chen, Jia-Horung Hung, Hsiao-Yen Chen, Chun-
Chieh Lai, Hon-Chun Cheng, Sung-Huei Tseng.
Formal analysis: Yi-Sheng Chang, Po-Lin Chen, Hsiao-Yen Chen, Chun-Yen Ou, Sung-Huei
Tseng.
Investigation: Yi-Sheng Chang, Po-Lin Chen, Jia-Horung Hung, Hsiao-Yen Chen, Chun-
Chieh Lai, Chun-Yen Ou, Chia-Ming Chang, Chien-Kuo Wang, Sung-Huei Tseng.
Methodology: Yi-Sheng Chang, Po-Lin Chen, Hsiao-Yen Chen, Chun-Chieh Lai, Chun-Yen
Ou, Chia-Ming Chang, Chien-Kuo Wang, Hon-Chun Cheng, Sung-Huei Tseng.
Project administration: Yi-Sheng Chang, Jia-Horung Hung, Sung-Huei Tseng.
Resources: Yi-Sheng Chang, Sung-Huei Tseng.
Software: Yi-Sheng Chang.
Supervision: Yi-Sheng Chang, Sung-Huei Tseng.
Validation: Yi-Sheng Chang, Chien-Kuo Wang, Sung-Huei Tseng.

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 12 / 14

 Orbital complications of sinusitis

Visualization: Yi-Sheng Chang, Chien-Kuo Wang, Sung-Huei Tseng.

Writing – original draft: Yi-Sheng Chang, Hsiao-Yen Chen, Chia-Ming Chang.
Writing – review & editing: Yi-Sheng Chang, Po-Lin Chen, Jia-Horung Hung, Hsiao-Yen
Chen, Chun-Chieh Lai, Chun-Yen Ou, Chia-Ming Chang, Chien-Kuo Wang, Hon-Chun
Cheng, Sung-Huei Tseng.

References
1. Bailey BJ. Head and neck surgery—otolaryngology. 4th ed. Philadelphia: Williams & Wilkins; 2001.
2. Hastan D, Fokkens WJ, Bachert C, Newson RB, Bislimovska J, Bockelbrink A, et al. Chronic rhinosinu-
sitis in Europe—an underestimated disease. A GALEN study. Allergy. 2011; 66: 1216–1223. https://doi.
org/10.1111/j.1398-9995.2011.02646.x PMID: 21605125
3. Shi JB, Fu QL, Zhang H, Cheng L, Wang YJ, Zhu DD, et al. Epidemiology of chronic rhinosinusitis:
results from a cross-sectional survey in seven Chinese cities. Allergy. 2015; 70: 533–539. https://doi.
org/10.1111/all.12577 PMID: 25631304
4. Chandler JR, Langenbrunner DJ, Stevens ER. The pathogenesis of orbital complications in acute sinus-
itis. Laryngoscope. 1970; 80: 1414–1428. https://doi.org/10.1288/00005537-197009000-00007 PMID:
5470225
5. Chaiyasate S, Fooanant S, Navacharoen N, Roongrotwattanasiri K, Tantilipikorn P, Patumanond J.
The complications of sinusitis in a tertiary care hospital: types, patient characteristics, and outcomes. Int
J Otolaryngol. 2015; 2015: 709302. https://doi.org/10.1155/2015/709302 PMID: 25722725
6. Dankbaar JW, van Bemmel AJ, Pameijer FA. Imaging findings of the orbital and intracranial complica-
tions of acute bacterial rhinosinusitis. Insights Imaging. 2015; 6: 509–518.
7. Howe L, Jones NS. Guidelines for the management of periorbital cellulitis/abscess. Clin Otolaryngol
Allied Sci. 2004; 29:725–728. https://doi.org/10.1111/j.1365-2273.2004.00889.x PMID: 15533168
8. Sobol SE, Marchand J, Tewfik TL, Manoukian JJ, Schloss MD. Orbital complications of sinusitis in chil-
dren. J Otolaryngol. 2002; 31: 131–136. PMID: 12121013
9. Ferguson MP, McNab AA. Current treatment and outcome in orbital cellulitis. Aust N Z J Ophthalmol.
1999; 27: 375–379. PMID: 10641894
10. Batra PS, Han JK. Practical medical and surgical management of chronic rhinosinusitis. 1st ed. Swit-
zerland: Springer; 2015.
11. Mortimore S, Wormald PJ. The Groote Schuur hospital classification of the orbital complications of
sinusitis. J Laryngol Otol. 1997; 111: 719–723. PMID: 9327008
12. Kinis V, Ozbay M, Bakir S, Yorgancilar E, Gun R, Akdag M, et al. Management of orbital complications
of sinusitis in pediatric patients. J Craniofac Surg. 2013; 24: 1706–1710. https://doi.org/10.1097/SCS.
0b013e3182a210c6 PMID: 24036759
13. Gupta S, Goyal R, Gupta RK. Clinical presentation and outcome of the orbital complications due to
acute infective rhino sinusitis. Indian J Otolaryngol Head Neck Surg. 2013; 65: 431–434. https://doi.org/
10.1007/s12070-013-0646-6 PMID: 24427692
14. Root RK. Clinical infectious diseases: A practical approach. New York: Oxford University Press; 1999.
15. Rubin SE, Slavin ML, Rubin LG. Eyelid swelling and erythema as the only signs of subperiosteal
abscess. Br J Ophthalmol. 1989; 73: 576–578. PMID: 2758001
16. Younis RT, Anand VK, Davidson B. The role of computed tomography and magnetic resonance imaging
in patients with sinusitis with complications. Laryngoscope. 2002; 112: 224–229. https://doi.org/10.
1097/00005537-200202000-00005 PMID: 11889374
17. Harris GJ. Subperiosteal abscess of the orbit. Arch Ophthalmol. 1983; 101: 751–757. PMID: 6847463
18. Arjman EM, Lusk RP, Muntz HR. Pediatric sinusitis and subperiosteal orbital abscess formation: diag-
nosis and treatment. Otolaryngol Head Neck Surg. 1993; 109: 886–894. https://doi.org/10.1177/
019459989310900518 PMID: 8247570
19. Harris GJ. Age as a factor in the bacteriology and response to treatment of subperiosteal abscess of the
orbit. Trans Am Ophthalmol Soc. 1993; 91: 441–516. PMID: 8140703
20. Peña MT, Preciado D, Orestes M, Choi S. Orbital complications of acute sinusitis: changes in the post-
pneumococcal vaccine era. JAMA Otolaryngol Head Neck Surg. 2013; 139: 223–227. https://doi.org/
10.1001/jamaoto.2013.1703 PMID: 23429877

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 13 / 14

 Orbital complications of sinusitis

21. Sharma S, Josephson GD. Orbital complications of acute sinusitis in infants: a systematic review and
report of a case. JAMA Otolaryngol Head Neck Surg. 2014; 140: 1070–1073. https://doi.org/10.1001/
jamaoto.2014.2326 PMID: 25317572
22. Stammberger H. Endoscopic endonasal surgery—concepts in treatment of recurring rhinosinusitis.
Part II. Surgical technique. Otolaryngol Head Neck Surg. 1986; 94: 147–156. https://doi.org/10.1177/
019459988609400203 PMID: 3083327
23. Garcia GH, Harris GJ. Criteria for nonsurgical management of subperiosteal abscess of the orbit: analy-
sis of outcomes 1988–1998. Ophthalmology. 2000; 107: 1454–1456. PMID: 10919887
24. Souliere CR Jr, Antoine GA, Martin MP, Blumberg AI, Isaacson G. Selective non-surgical management
of subperiosteal abscess of the orbit: computerized tomography and clinical course as indication for sur-
gical drainage. Int J Pediatr Otorhinolaryngol. 1990; 19: 109–119. PMID: 1695617
25. Patt BS, Manning SC. Blindness resulting from orbital complications of sinusitis. Otolaryngol Head
Neck Surg. 1991; 104: 789–795. https://doi.org/10.1177/019459989110400604 PMID: 1908969
26. Rubinstein JB, Handler SD. Orbital and periorbital cellulitis in children. Head Neck Surg. 1982; 5: 15–
21. PMID: 7174342

PLOS ONE | https://doi.org/10.1371/journal.pone.0184477 October 3, 2017 14 / 14